More than one million men have been unnecessarily treated for prostate cancer due to the widespread use of a common screening test that detects early markers of cancer cells that may never develop into malignant tumors, according to a study co-authored by Dartmouth Medical School professor H. Gilbert Welch, director of the Center for Medicine and the Media at The Dartmouth Institute for Health Policy and Clinical Practice.
In many patients diagnosed with prostate cancer, the early cancerous growth found by the prostate-specific antigen commonly known as PSA screening test may not pose an actual health risk, the study found. This leads to overdiagnosis, inflating the number of prostate cancer cases, and unnecessary treatment.
Welch's study, published online in the Journal of the National Cancer Institute on Aug. 31, was co-authored with University of Connecticut School professor Peter Albertsen.
Analyzing data from 1986, the year before the PSA test was introduced, to 2005, Welch and Albertson found that an additional 1,305,600 men were diagnosed with prostate cancer than would be expected based on the assumption that the underlying prevalence of prostate cancer did not change. About one million of these men subsequently received treatment for the disease. Many of these additional incidences of prostate cancer, the researchers suggested, were likely the result of overdiagnosis.
Welch acknowledged, however, that overdiagnosis is generally difficult to measure at the individual level, as diagnosis of cancer usually results in intervention.
"The weakness of overdiagnosis is a little like black holes in physics," he said. "You can't directly observe the phenomenon. You have to make inferences about it."
The PSA test involves measuring the amount of prostate-specific antigen, a substance produced by prostate gland cells, from a blood sample. Abnormal PSA levels may indicate prostate cancer, but they can also reflect benign tumors or other prostate conditions. If elevated levels of PSA are detected, other tests may be performed to determine if cancer is present.
"In the ideal world, screening wouldn't influence how many people were said to have cancer. It would only influence the time at which [the cancer was diagnosed]," Welch said in an interview with The Dartmouth.
The researchers estimated that about 56,500 prostate cancer deaths were avoided due to use of the PSA test, assuming that the total decrease in deaths over the period was due to use of the test.
Prostate cancer treatment often involves removing the prostate or initiating radiation therapy. These procedures, Welch said, can lead to sexual dysfunction, incontinence or urinary dysfunction in about one-third of cases.
"If the cancer was never going to bother you and cause symptoms, the only thing treatment can do is hurt you," he said.
Early diagnosis of prostate cancer may also have psychological effects, he said, describing unnecessary diagnosis as "extremely anxiety-producing."
"The whole issue is more nuanced than is commonly appreciated, that screening is a two-edged sword," he said. "Yes, it may help men, but it may also hurt them."
The American Cancer Society does not recommend routine testing for prostate cancer, though it suggests that physicians offer the PSA test, in addition to another test, to men age 50 and older. The American Urological Association recommends that physicians begin offering the test to men age 40 and older. Both acknowledge that there are significant shortcomings associated with early cancer detection.
